OSI's war on corporate licenses

Tue Apr 12 22:55:51 UTC 2005

"war on corporate licenses"? I don't know. I have a different take. I
see a reason to apply more effort to reduce the number of different
copyleft licenses than the number of non-copyleft licenses. 

In general, code under a copyleft license cannot be blended code under
other copyleft licenses. 

In contrast, code under any non-copyleft license can be blended with
code under any other non-copyleft license (and some of the most popular
will even blend with copyleft licenses). 

A large number of licenses of any type adds complexity. However,
proliferation of copyleft licenses has the added disadvantage that they
create separate pools of unmixable or semi-mixable code - an impediment
that is not created by proliferation of non-copyleft licenses. 

License proliferation can be the cause of two separate types of
inhibition that I believe are worth considering separately:

License proliferation creates complexity - often gratuitously.
Proliferation of long licenses creates more complexity than
proliferation of short licenses. Complexity can be an inhibitor that we
ought to seek to reduce. 

License proliferation can impede code reuse. This is an issue unique to
proliferation of copyleft licenses. 

Thus, I believe there is value in paying somewhat more attention to the
proliferation of copyleft licenses than to proliferation of non-copyleft
licenses. 

The licenses generally identified as 'corporate' are often long and
often include a copyleft feature. Thus, there is a high correlation
between corporate licenses and the licenses that I think worthy of
particular attention when trying to reduce impediments flowing from
license proliferation. 

-- Scott
______________________________
Scott K. Peterson
Senior Counsel
Hewlett-Packard Company
scott.k.peterson at hp.com

-----Original Message-----
From: Joel West [mailto:svosrp at gmail.com] 
Sent: Monday, April 11, 2005 9:23 PM
To: license-discuss at opensource.org
Subject: OSI's war on corporate licenses

The list has discussed the problem of the proliferation of "vanity"
licenses. But it seems the OSI board has gone far beyond license
proliferation, to actively discourage firm-sponsored OSS licenses of any
kind.

At the new policy
	http://opensource.org/docs/policy/licenseproliferation.php 
I found this statement troubling:

>The class of asymmetrical corporate licenses that began with Mozilla
was a
>worthy experiment that has failed. The new policy will discourage them.

and particularly

> But seven years later, we think it is is significant that the original
> corporate open-source license, the Mozilla Public License, has been
dropped
> by its originating organization in favor of the GPL. It is becoming
> increasingly clear from this and other examples that the "middle way"
> represented by Mozilla and other corporate open-source licenses is not
> a stable, effective solution even from the point of view of selfish
> corporate agents.

I don't know anything more than what's on these webpages. But from
reading these statements, it's not clear what planet the authors are on.

Mozilla is no longer a corporate-sponsored OSS project, but instead is
an independent stand-alone project that in most respects is
indistinguishable from projects that began as community-founded projects
(e.g. Apache). So since they are organized more like Apache or Project
GNU, it's not surprising that they have a community license. All it
proves is that AOL gave up on selling browsers.

>These licenses put a hard brake on the growth of development 
>communities around their products without actually delivering 
>measurable advantages in revenue, market control, or risk 
>management. Because these licenses have largely failed to deliver, 

Were any of these people at EclipseCon? There's a lot of development,
community, revenue and adoption going on in Eclipse as the .NET
alternative. It's all based on the CPL (soon to be EPL), a slightly
cleaned up MPL. New strategic developers were announced including a good
portion of the major ISVs in the tools area (with the notable exception
of Sun and Microsoft).

So has Eclipse "failed to deliver"?

As Evan said, there is a problem with license incompatibility. This may
be as much a problem with the conditions of the GPL as it is with the
problem of the other licenses. And maybe Martin Fink is right, there's a
problem with license proliferation.

But how does this lead to the conclusion that firm-sponsored OSS
licenses or projects are a failure? There's a logical leap that's been
left out.

Or is this just an attack on asymmetric open source business models? And
if we're going to attack asymmetric models, why the MPL/CPL/CDDL? Why
not attack the dual license GPL model? In both cases, corporations that
object to viral (aka copyleft) provisions can pay to escape them. In
both cases, big bad corporations own the IPR rather than the broader
community.

We are just beginning to see a huge experimentation in open source
business models. It's hard to see how OSI or the open source movement
are helped if those involving dual license get to call themselves "open
source" but those involving the CPL do not. For that matter, if OSI
kicks out the CPL crowd (and/or GNU kicks out the dual license crowd),
we might have a new movement formed with Sun, IBM, HP, Microsoft and
others. And if that's where all the investment is going, it's hard to
see how that helps OSS or OSI at all.

Disclaimer: I teach in a business school and believe in market
incentives to promote maximum innovation and investment. I think the
original 1998 group (immortalized by "Revolution OS") was brilliant in
leveraging this principal to create the Open Source movement.

Joel

-- 
Joel West, Research Director
Silicon Valley Open Source Research Project
http://www.cob.sjsu.edu/OpenSource/